Local Frequency Attenuation Gradient Of Seismic Wave In Special Reservoir Analysis

Special reservoirs include carbonate reservoir, volcanic reservoir, densesandstone reservoir etc. Currently, oil and gas resources in the special reservoirs arethe most realistic alternative energy sources and play an increasingly important role inthe structure of the world’s energy. In China, with speed up of the industrializationprocess, energy demand is also growing rapidly, energy supply, energy environment,sustainable development, and other issues have increasingly become prominent. In2011, the total domestic production of crude oil are204million tons, an increase of0.3percent over the previous year; imported a total of254million tons of crude oil,an increase of6%over the previous year; imported a total of40.6million tons ofrefined oil, an increase of10.1percent over the previous year. Dependence on foreignoil reached56.5%. China has become the world’s second largest oil importer andconsumer country after the United States. The energy issue is a big problemrestricting China’s sustainable development, in order to tackle this problem, Chinaneeds to vigorously carry out a plan to develop and utilize oil and gas resources inspecial reservoir.More high accuracy exploration is required with deeper exploration. How toidentify oil and gas effectively, thereby reducing exploration risk is a thorny problemin front of the oil and gas exploration workers. Since many years some geophysicsworkers have devoted themselves to the research about it. Among them theapplication of attenuation properties of the seismic signal frequency is becominghotspot. Generally speaking, oil or gas is largely relevant to frequencies attenuation,but because of the complicated underground media, wave equation is barely tonumerically model the seismic wave travel. The developed frequency attenuationproperties are not able to fully apply to all exploration work area, so this urges thegeophysics workers to continue to make their efforts in the future.This paper focuses on the study about the frequency attenuation of seismic wavesin the special reservoir. Firstly summarize the lithological and geophysicalcharacteristics of carbonate and volcanic reservoirs. Then introduce the conventionaltime-frequency analysis, time-frequency decomposition mapping a1D signal into a 2D time and frequency domain, and describe how the spectral content of the signalchanges with time. Time-frequency analysis has been used extensively in seismic dataprocessing and interpretation, including attenuation measurement and stratigraphicmapping. The widely used short-time Fourier transform（STFT）method produces atime-frequency spectrum by taking the Fourier transform of data windows, whichleads to a tradeoff between temporal and spectral resolution.Over the past two decades, wavelet-based methods have been applied totime-frequency analysis of seismic data. Former researchers compare wavelet-basedwith Fourier-based methods for performing time-frequency analysis on seismic dataand show that the wavelet-based method improves spectral resolution. Thecontinous-wavelet transform（CWT）provides a time-scale map, known as scalogramrather than a time-frequency spectrum. Because of a scale represents a frequency band,we can choose their scale to be inversely proportional to the center frequency of thewavelet, which allowed them to transform their scalogram into a time-frequency map.S transform provides a novel method of obtaining such a time-frequency map bytaking a Fourier transform of the inverse CWT. Although S transform has its merits,but its own large amount of calculation, the slower computing speed and relativelypoor resolution features limit its application in the analysis of the seismic dataattributes.Therefore, this paper introduce local time-frequency transform, localtime-frequency transform is different from short time Fourier transform and Stransform, because local time-frequency transform does not require the localtime-window selection, but a calculation by employing the regularization parameterinto the time-frequency parameter. In order to solve the underdeterminedmathematical problem of the least squares estimation, we introduce the shapingregularization. Shaping regularization provides a particularly convenient method ofenforcing smoothness in iterative optimization schemes. In this paper, we use shapingregularization, analogous to the problem of nonstationay regression, to constrainestimated coefficient. Local time-frequency transform apply to the self-adaptiveanalysis of actual data in the time-frequency domain by controlling the parameters ofthe smoothing radius.After introduce and compare several time-frequency analysis methods, in thispaper, we analyze the attenuation theory of seismic waves and seismic wavesfrequency. And then use of two point methods of fitting a straight line to getfrequency attenuation gradient in order to analysis the trend of high frequency attenuation. The frequency attenuation gradient based on short time Fourier transform,S transform and local time-frequency transform can reflect the high frequencyattenuation in actual volcanic seismic reservoir.